Massive frozen osteochondral allografts are currently being used for reconstructive orthopaedic surgery following limb-sparing en bloc tumor resections, segmental skeletal losses resulting from trauma and for replacement of arthritic joint surfaces. While initial clinical impressions are encouraging, the ultimate success of this approach depends upon identification of biologic, immunologic and biomechanical parameters most compatible with individual recipient needs and a system of banking safe and efficacious allografts of sufficient variety and quantity to support the clinical program. Sensitive and objective histomorphometric techniques will be used to evaluate the biology of repair and incorporation as reflected in cortical and cancellous test aliquots of bone implanted at the time of primary reconstructive surgery and available for retrieval 12 weeks later in patients being treated at the collaborating institutions. The massive grafts (and test aliquots) will be dispensed from a uniform multi-institution banking program with tissues being chosen on the basis of biomechanical considerations alone, particularly fit at the articular surfaces. Tissue typing (A, B and DR) of donors and recipients will be determined but not be intentionally matched. Post-operative antibody screens for anti-HLA responses will be conducted at 3 month intervals during at least the first 2 years of follow-up. Clinical success will be measured by evaluating function, x-rays and technetium scans. Correlation of clinical success with biological changes (histomorphometric analysis), biochemical consideration (joint congruity and stable fixation) and immunological responses will identify those parameters predictive of clinical success and provide a basis for determining appropriate tissue matches, banking methods and transplantation techniques on a prospective basis in the future.